The beginning of intelligence - a place to begin the search!

I believe we need to start looking at the moment an organism was able to both control its movement and recognize a pattern.  I don't think this would be too difficult to set up.  You would use food, much like experiments with mice.  I am assuming if food is positioned in straight lines, an organism would go in straight lines.  Then I am assuming the organism would continue to search in a straight line if giving the food was stopped.  The experimenter would then place small bits of food back in the opposite direction, eventually giving the "motherload" of food.  This process would be repeated x amount of times, probably at least 100.  After the x amount of trials with the food trail, the next trial would not have the food trail leading the organism back.  If the organism does go back to the designated spot, the "motherload" would be given.  If an organism given this training repeats it consistently, that would be evidence that learning a pattern has occurred, and thus intelligence!

I would begin with organisms known to be able to "turn off" their flagella, possibly protists.